基于标签和共评比例改进的推荐算法  

Improved Recommendation Algorithm Based on Label and Co-evaluation Ratio

作　　者：林鹏飞 王逊[1] 黄树成[1] LIN Pengfei;WANG Xun;HUANG Shucheng(School of Computer Science and Engineering,Jiangsu University of Science and Technology,Zhenjiang 212003)

机构地区：[1]江苏科技大学计算机科学与工程学院,镇江212003

出　　处：《计算机与数字工程》2023年第9期2067-2073,共7页Computer & Digital Engineering

基　　金：国家自然科学基金项目(编号:61772244)资助。

摘　　要：针对数据稀疏性问题,提出了一种新的相似度计算方法来提高传统协同过滤方法(CF)的精度。根据与用户有强相关性的用户偏好进行分析,向用户提供他们所需的项目。皮尔逊相关系数和余弦相似度,作为应用最广泛的方法,仅根据用户对项目的共同评分来发现用户之间的相关性。因此,这些方法缺乏解决稀疏性的能力。论文提出了一种新的基于全局用户偏好的相似度方法来解决稀疏性问题,提高推荐的准确性。因此,该方法的新颖之处在于能够解决相似性问题,同时能够发现不相关用户之间的关系。此外,在计算一对用户之间的相似度的过程中,为了确定正确的邻居数量,该方法考虑了两个主要因素(公平性和共评比例)。并在MovieLens 100K数据集下用于评估论文算法的准确性。实验结果表明,与传统CF相似性方法相比,该方法在各项指标上都有所提高。Aiming at the problem of data sparsity,a new similarity calculation method is proposed to improve the accuracy of the traditional collaborative filtering method(CF).It analyzes based on user preferences that are strongly relevant to users,and provides users with the items they need.Pearson's correlation coefficient and cosine similarity,as the most widely used method,only find the correlation between users based on the users'common ratings of items.Therefore,these methods lack the ability to solve sparsity.This paper proposes a new similarity method based on global user preferences to solve the sparsity problem and improve the accuracy of recommendation.Therefore,the novelty of this method is that it can solve the similarity problem and discover the relationship between unrelated users.In addition,in the process of calculating the similarity between a pair of users,in order to determine the correct number of neighbors,the method considers two main factors(fairness and co-evaluation ratio).And it is used to evaluate the accuracy of the algorithm in this paper under the MovieLens 100K data set.The experimental results show that compared with the traditional CF similarity method,this method has improved various indicators.

关 键 词：协同过滤 用户偏好 用户相似度 公平性 共评比例 

分 类 号：TP181[自动化与计算机技术—控制理论与控制工程]